Diabetes currently affects 55.8 million people, or 8.3% of the U.S. population, with prevalence in Hispanics particularly high at 11.8%. Several studies have shown significant associations of diabetes with persistent organic pollutants (POPs), including organochlorine pesticides, polychlorinated biphenyls (PCBs), and brominated flame retardants (BFRs). The biologic pathways by which these exposures could increase risk of diabetes have not been elucidated, although there is evidence that inflammatory and endocrine mechanisms may be involved. Use of chlorinated pesticides continued longer in Mexico and other selected Latin American countries and levels are higher in Hispanics in the US than in other ethnic groups. The proposed study will build upon the Hispanic Community Health Study/Study of Latinos (HCHS/SOL), an ongoing cohort of 16,415 multiethnic Hispanics in 4 cities in the US: Chicago, San Diego, New York and Miami. Baseline bloods samples from 2,850 men and postmenopausal women age 45-74 will be selected for measurements of 9 organochlorine pesticides, 11 BFRs, and 37 PCBs, as well as endogenous thyroid and steroid hormones and measures of inflammation that have been related to the exposures and/or diabetes. Relationships of levels of POPs, endogenous hormones and measures of inflammation with demographic factors (including age, gender Hispanic background group and years of residence in the US), and baseline diabetes, pre-diabetes and insulin resistance will be examined. Associations of baseline levels of POPs, measures of inflammation and endogenous hormones with subsequent development of diabetes, pre-diabetes and insulin resistance determined at the six year follow-up will be examined. Modification of associations of POPs with diabetes, pre-diabetes and insulin resistance by gender, age, body mass index, ethnicity, immigration status, hormones, diet and inflammation markers will be examined with interaction terms and stratification. Effects of potential intermediate markers such as hormones and measures of inflammation will be explored with mediation analysis. The HCHS/SOL Study is the largest prospective study in the United States of Hispanic groups of diverse background. The wide variation in both exposures and disease in this population renders it ideal in which to examine biologic pathways by which POPs could be affecting an increasingly prevalent and costly disease. Hypotheses developed in this model can later be extrapolated to other common exposures that are less persistent and more difficult to study. Results could be important not only in identifying etiologic factors, but also in designing appropriate intervention strategies.